This SBIR Phase I proposal is to generate novel tumor necrosis factor (TNF) dominant negative variants that will inactivate the activity of TNF receptors. The dominant negative TNF monomers will still selectively bind to other wild type TNF monomers forming a heterotrimer complex thereby sequestering normal TNF trimer activity. Furthermore, these TNF heterotrimer complexes are engineered to also not bind to the TNF-receptor homotrimer precluding the downstream signaling activity and inflammatory responses associated with Rheumatoid arthritis. We will employ Protelix's proprietary Look Through Mutagenesis (LTM) to systematically mutate the receptor contact regions in the TNF protein for accelerated and efficient generation of targeted variant libraries. Subsequent use of microbial surface display system then allows for high-throughput screening of our TNF variant library to find clones with differential binding affinities. Each screen recovers informational gains on beneficial mutations whereupon the reiterative process of combining beneficial mutations (CBM) is then applied to progressively optimize new multiple-site variants. The best dominant negative TNF clones are then selected from standard kinetic and in vitro bioassays as possible therapeutics for Phase II development. This SBIR Phase I proposal is to produce new versions of tumor necrosis factor (TNF) that interfere with the natural TNF in the body. Rheumatoid arthritis one of the biological consequences of excess TNF can be eliminated or greatly reduced by selectively blocking the action of normal TNF. We will use Protelix's technology to develop variants of tumor necrosis factor by making them bind to normal TNF but not to the TNF receptor complex involved in inflammation. We will then test the best tumor necrosis factor variants in a variety of kinetic measures, cell-based assays for superior activity in preventing cell death before we test their protective biological efficacy in animal models. [unreadable] [unreadable] [unreadable]